A method based on rf spectral featuresfor evaluating the porosity degree in ceramic materials

Abstract

In this paper, a classification system of the degree of porosity of ceramic materials based on a Radio Frequency system is presented. The system uses methods from the machine learning field to learn patterns from spectral features measured with a circular patch antenna. Experimental results show that it is possible to indirectly get an estimate of the degree of porosity of ceramic samples getting low classification error rates.